The research program is focused on the detailed mechanisms underlying the initiation of innate immune responses by the receptors and the ensuing signal transduction pathways. The innate immune system is the first line of defense against infection that also initiates and directs the proper function of the adaptive immune response. The potency of the innate immune system has been harnessed as vaccination adjuvants against infections, cancer, and autoimmune diseases. Nonetheless, the underlying mechanisms of action only started to be delineated about ten years ago, when a major family of the pattern recognition receptors (PRRs) was identified. It is now appreciated that several families of PRRs cross-regulate each other in coordinating innate immune responses. Recent crystal structures of the extracellular domains of TLRs have provided insights on the ligand recognition by different receptors, demonstrating that the mechanisms of ligand recognition varies significantly among different PRRs. Our program integrates biochemical studies with biophysical characterization of innate receptors such as AIM2, IFI16, NLRP1, RIG-I and TLR9, either in complex with their ligands or downstream adapters/effector molecules. A critical feature of these innate immune receptors is that they distinguish among various classes of pathogenic molecules while retaining their capacity for responsiveness to a large number of related structures within a given biochemical class. How the ligand-binding domains of the innate receptors achieve such broad reactivity at the atomic level is one of the key issues this project addresses. Such information could be used to guide the development of new therapeutics that can either enhance or limit immune activation involving these receptors. Ligand binding by these receptors in turn initiates molecular signaling cascades that ultimately lead to innate cellular responses that help fight infection and guide the adaptive immune responses. The project aims to decipher this signaling network through studying protein-protein interactions, using X-ray crystallography in conjunction with other biophysical and biochemical techniques. The ultimate goal is to not only delineate the mechanisms of innate immune signal transduction at atomic details, but also to lay a foundation for future clinical exploitation of the innate immune system, such as the development of more effective vaccine adjuvants. During 2009-2010 fiscal year, we have achieved progress in the following area: 1). Through modification of MyD88 TIR domain using Lysine methylation, we first determined its crystal structure. Since then, while screening crystallization of MyD88 TIR domain in complex with the TIR domain of TcpC, a pathogen protein that targets MyD88 signaling, we obtained crystals of un-modified MyD88 TIR domain at much higher resolution of 1.5 . The structure shows almost identical fold and crystal lattice contacts, validating our previous observation of MyD88 TIR domain structure. 2). Recently we have crystallized the TIR domains of human TIRAP, a host partner protein for MyD88. We have solved the structure of the human TIRAP TIR domain at 2.6 resolution using a seleno-methionine MAD dataset. The structure shows unexpected conformational differences compared with the classic TIR domains, with one-less &#945;helix and one-less &#946;strand. Most strikingly, residues around the predicted BB-loop are completely disordered with minimal observable electron density, and several residues at the predicted helix C adopt &#946;strand structure to shield the hydrophobic core. In order to understand such drastic structural differences, we are carrying out NMR titration studies of human TIRAP TIR domain to investigate its biophysical properties under solution conditions. 3). TcpB is a bacterial protein from Brucella that interacts with TIRAP to target host innate immune system. It has been shown that both TcpB ad TIRAP binds lipid molecules and locate to membrane surface on the cytosolic side, which may facilitate their interaction. In addition, phosphorylation of TIRAP and/or TcpB appears to modulate their interaction and degradation. In collaboration with our colleagues from the Miethke lab at Germany, we successfully expressed and purified TcpB, and crystallized its TIR domain. The crystals diffracted very well using home-source X-ray data collection system to 2.5 resolution. We have just solved its structure and are carrying out refinement of the model against the X-ray diffraction data. The current structure shows a typical TIR domain fold with symmetric dimer formation. Characterization of the TcpB TIR domain surface properties and docking onto the structure of TIRAP is currently ongoing. 4). Concurrently, we are continuing studies of TIR-TIR domain interactions using NMR titration experiments in collaboration with Nico Tjandras group at NHLBI. Our NMR titration experiments illustrated the surface of MyD88 that interacts with TcpC. We are carrying out the reverse titration to locate the interface on the TcpC side. We will carry out similar titration experiments with the TIRAP-TcpB interacting TIR domain pair. We will further correlate these TIR-TIR binding with our yeast two-hybrid results. 5). Recently we obtained crystals for the CTD domain of RIG-I. The structure is identical to the published crystal and NMR structures, with a conserved crystal lattice interface mediated by a continuous &#946;sheet across neighboring molecules. The packing interaction is present in the previously published crystal structure but was not described in the accompanying publication. We are looking into the possibility that this packing interaction may facilitate oligomerization of the full-length RIG-I. 6). The challenges of obtaining crystals for full-length RIG-I prompted us to use alternative biophysical techniques to study its structure. Using small-angle X-ray scattering (SAXS), we have investigated global conformational changes of the full-length or near full-length RIG-I upon ligand binding. Our initial results indicate that full-length RIG-I exists as a spoon-shaped monomer in solution, and its size (radius of gyration) increases with the binding of ligand (double-stranded/ds RNA), suggesting conformational changes. However, the increase in size does not support the ligand-induced dimerization model proposed in the literature. We are testing whether longer length of dsRNA may induce dimerization of RIG-I. In addition, we are looking into collecting SAXS data at lower temperature, as our previous data suggest that RIG-I forms aggregation upon exposure to X-ray at room temperature. This part of work is currently being carried out in collaboration with John Tainers group at the Scripps Research Institute. 7). The HIN-200 family of proteins was recently identified as a new family of cytosolic DNA receptors. The HIN domain has been known to bind DNA, and the Pyrin domain at the N-termini of this family of proteins appears to be the signaling domain that interacts with adapter ASC. Among this family of proteins, AIM2 (absent in melanoma 2) was shown to induce caspase-1 maturation and IL-1&#946;secretion through association with ASC. Recently we obtained crystals of the AIM2 and IFI16 HIN domains in complex with DNA, and have solved their structures that contain DNAs of various lengths. Our structures demonstrate that the HIN domains interact with the sugar-phosphate backbone of DNAs and make no contacts with the DNA bases, consistent with their roles as generic DNA sensor with no sequence/base specificity. We are carrying out surface plasmon resonance (SPR) studies of the DNA-HIN binding to further characterize their interactions, and confirm the HIN interface residues using wild-type and mutant HIN domains.